On Mar 26, 2007, at 9:28 PM, DJ Delorie wrote:
No, it's more like this:typedef int copsi __attribute__((mode(COPSI))); void foo (int *a, copsi *b, int i) { while (i--) { *a *= 2; *b *= 2; } } This will keep both the core multiplier and the coprocessor multiplier busy.
:-( Wouldn't it be better to eventually get the compiler to schedule and use all the resources on the chip by itself? I'm sure it is a larger job, but I'd hope that we're not unreasonably far from being able to do that?